Nonlinear interpolation fractal classifier for multiple cardiac arrhythmias recognition

Chia Hung Lin, Yi Chun Du, Tainsong Chen

Research output: Contribution to journalArticle

15 Citations (Scopus)

Abstract

This paper proposes a method for cardiac arrhythmias recognition using the nonlinear interpolation fractal classifier. A typical electrocardiogram (ECG) consists of P-wave, QRS-complexes, and T-wave. Iterated function system (IFS) uses the nonlinear interpolation in the map and uses similarity maps to construct various data sequences including the fractal patterns of supraventricular ectopic beat, bundle branch ectopic beat, and ventricular ectopic beat. Grey relational analysis (GRA) is proposed to recognize normal heartbeat and cardiac arrhythmias. The nonlinear interpolation terms produce family functions with fractal dimension (FD), the so-called nonlinear interpolation function (NIF), and make fractal patterns more distinguishing between normal and ill subjects. The proposed QRS classifier is tested using the Massachusetts Institute of Technology-Beth Israel Hospital (MIT-BIH) arrhythmia database. Compared with other methods, the proposed hybrid methods demonstrate greater efficiency and higher accuracy in recognizing ECG signals.

Original languageEnglish
Pages (from-to)2570-2581
Number of pages12
JournalChaos, solitons and fractals
Volume42
Issue number4
DOIs
Publication statusPublished - 2009 Nov 30

All Science Journal Classification (ASJC) codes

  • Statistical and Nonlinear Physics
  • Mathematics(all)
  • Physics and Astronomy(all)
  • Applied Mathematics

Fingerprint Dive into the research topics of 'Nonlinear interpolation fractal classifier for multiple cardiac arrhythmias recognition'. Together they form a unique fingerprint.

  • Cite this